This project aims to evaluate the utility of simulated longitudinal MRI scans generated by the deep-learning-based InBrainSyn framework for clinical use by radiologists and geriatricians, in people with Alzheimer’s disease (AD) and normal aging. Alzheimer’s disease is a leading cause of dementia and death in the geriatric population, with a growing need for innovative methods to track neurodegeneration. While MRI is widely used for diagnosing AD, obtaining longitudinal data is often challenging due to high costs and logistical constraints. InBrainSyn leverages advanced deep learning techniques, including a state-of-the-art deep deformable spatiotemporal template generation model, called Atlas-GAN (https://github.com/neel-dey/Atlas-GAN), to create realistic, subject-specific simulations at different ages from a single brain MRI scan. By integrating deep learning-based registration networks and stationary velocity field frameworks, InBrainSyn ensures precise alignment and synthesis of individual-level longitudinal transformations. The project has three main objectives: (1) generate high-resolution simulated MRI scans using the deep-learning-based InBrainSyn framework and the publicly-available OASIS3 dataset, (2) use a machine learning (ML) technique, called AVRA (https://github.com/gsmartensson/avra_public), and a more research-oriented software like FreeSurfer, to assess neurodegeneration in the simulated versus original scans. AVRA is an ML technique that automatically performs radiological assessments used in clinical settings to rate regional atrophy for each generated scan, and (3) to examine the association between MRI biomarkers such as brain atrophy in the simulated scans and cognitive decline in both AD patients and healthy elderly individuals. By validating these synthetic scans for clinical use, the current study seeks to offer an efficient approach for tracking brain changes over time, enhancing early diagnosis and personalized treatment strategies in the geriatric population.